Earnings Inequality and Mobility Trends in the United States: Nationally Representative Estimates from Longitudinally Linked Employer-Employee Data

95 Pages Posted: 19 Mar 2017 Last revised: 26 Feb 2019

See all articles by John M. Abowd

John M. Abowd

U.S. Census Bureau; Cornell University Department of Economics; Labor Dynamics Institute; School of Industrial and Labor Relations; NBER (on leave); CREST; IZA Institute of Labor Economics

Kevin L McKinney

University of California, Los Angeles (UCLA)

Nellie Zhao

Cornell University - Department of Economics

Multiple version iconThere are 2 versions of this paper

Date Written: March 2017

Abstract

Using earnings data from the U.S. Census Bureau, this paper analyzes the role of the employer in explaining the rise in earnings inequality in the United States. We first establish a consistent frame of analysis appropriate for administrative data used to study earnings inequality. We show that the trends in earnings inequality in the administrative data from the Longitudinal Employer-Household Dynamics Program are inconsistent with other data sources when we do not correct for the presence of misused SSNs. After this correction to the worker frame, we analyze how the earnings distribution has changed in the last decade. We present a decomposition of the year-to-year changes in the earnings distribution from 2004-2013. Even when simplifying these flows to movements between the bottom 20%, the middle 60%, and the top 20% of the earnings distribution, about 20.5 million workers undergo a transition each year. Another 19.9 million move between employment and non-employment. To understand the role of the firm in these transitions, we estimate a model for log earnings with additive fixed worker and firm effects using all jobs held by eligible workers from 2004-2013. We construct a composite log earnings firm component across all jobs for a worker in a given year and a non-firm component. We also construct a skill-type index. We show that, while the difference between working at a low- or middle-paying firm are relatively small, the gains from working at a top-paying firm are large. Specifically, the benefits of working for a high-paying firm are not only realized today, through higher earnings paid to the worker, but also persist through an increase in the probability of upward mobility. High-paying firms facilitate moving workers to the top of the earnings distribution and keeping them there.

Institutional subscribers to the NBER working paper series, and residents of developing countries may download this paper without additional charge at www.nber.org.

Suggested Citation

Abowd, John Maron and McKinney, Kevin L and Zhao, Nellie, Earnings Inequality and Mobility Trends in the United States: Nationally Representative Estimates from Longitudinally Linked Employer-Employee Data (March 2017). NBER Working Paper No. w23224. Available at SSRN: https://ssrn.com/abstract=2935429

John Maron Abowd (Contact Author)

U.S. Census Bureau ( email )

4600 Silver Hill Road
Washington, DC 20233
United States
+1.301.763.5880 (Phone)

Cornell University Department of Economics ( email )

261 Ives Hall
Ithaca, NY 14853-3901
United States

HOME PAGE: http://www.economics.cornell.edu

Labor Dynamics Institute ( email )

Ithaca, NY 14853-3901
United States

HOME PAGE: http://www.ilr.cornell.edu/LDI/

School of Industrial and Labor Relations ( email )

Ithaca, NY 14853-3901
United States

HOME PAGE: http://www.ilr.cornell.edu/LDI/

NBER (on leave) ( email )

1050 Massachusetts Avenue
Cambridge, MA 02138
United States

CREST ( email )

92245 Malakoff Cedex
France

HOME PAGE: http://www.crest.fr/

IZA Institute of Labor Economics

P.O. Box 7240
Bonn, D-53072
Germany

Kevin L McKinney

University of California, Los Angeles (UCLA) ( email )

405 Hilgard Avenue
Box 951361
Los Angeles, CA 90095
United States

Nellie Zhao

Cornell University - Department of Economics ( email )

414 Uris Hall
Ithaca, NY 14853-7601
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
21
Abstract Views
208
PlumX Metrics